Fungal infections of humans range from superficial conditions, usually caused by dermatophytes or Candida species, that affect the skin (such as dermatophytoses) to deeply invasive and often lethal infections (such as candidiasis and cryptococcosis). Pathogenic fungi occur worldwide, although particular species may predominate in certain geographic areas.
In the past 20 years, fungal infections have increased dramatically--along with the numbers of potentially invasive species. Indeed, fungal infections, once dismissed as a nuisance, have begun to spread so widely that they are becoming a major concern in hospitals and health departments. Fungal infections occur more frequently in people whose immune system is suppressed (because of organ transplantation, cancer chemotherapy, or the human immunodeficiency virus), who have been treated with broad-spectrum antibacterial agents, or who have been subject to invasive procedures (catheters and prosthetic devices, for example). Fungal infections are now important causes of morbidity and mortality of hospitalized patients: the frequency of invasive candidiasis has increased tenfold to become the fourth most common blood culture isolate (Pannuti et al. (1992) Cancer 69:2653). Invasive pulmonary aspergillosis is a leading cause of mortality in bone-marrow transplant recipients (Pannuti et al., supra), while Pneumocystis carinii pneumonia is the cause of death in many patients with acquired immunodeficiency syndrome in North America and Europe (Hughes (1991) Pediatr Infect. Dis J. 10:391). Many opportunistic fungal infections cannot be diagnosed by usual blood culture and must be treated empirically in severely immunocompromised patients (Walsh et al. (1991) Rev. Infect. Dis. 13:496).
The fungi responsible for life-threatening infections include Candida species (mainly Candida albicans, followed by Candida tropicalis), Aspergillus species, Cryptococcus neoforms, Histoplasma capsulatum, Coccidioides immitis, Pneumocystis carinii and some zygomycetes. Treatment of deeply invasive fungal infections has lagged behind bacterial chemotherapy.
There are numerous commentators who have speculated on this apparent neglect. See, for example, Georgopapadakou et al. (1994) Science 264:371. First, like mammalian cells, fungi are eukaryotes and thus agents that inhibit fungal protein, RNA, or DNA biosynthesis may do the same in the patient's own cells, producing toxic side effects. Second, life-threatening fungal infections were thought, until recently, to be too infrequent to warrant aggressive research by the pharmaceutical industry. Other factors have included:
(i) Lack of drugs. A drug known as Amphotericin B has become the mainstay of therapy for fungal infection despite side effects so severe that the drug is known as "amphoterrible" by patients. Only a few second-tier drugs exist. PA1 (ii) Increasing resistance. Long-term treatment of oral candidiasis in AIDS patients has begun to breed species resistant to older anti-fungal drugs. Several other species of fungi have also begun to exhibit resistance. PA1 (iii) A growing list of pathogens. Species of fungi that once posed no threat to humans are now being detected as a cause of disease in immune-deficient people. Even low-virulence baker's yeast, found in the human mouth, has been found to cause infection in susceptible burn patients. PA1 (iv) Lagging research. Because pathogenic fungi are difficult to culture, and because many of them do not reproduce sexually, microbiological and genetic research into the disease-causing organisms has lagged far behind research into other organisms.
In the past decade, however, more antifungal drugs have become available. Nevertheless, there are still major weaknesses in their spectra, potency, safety, and pharmacokinetic properties, and accordingly it is desirable to improve the the panel of anti-fungal agents available to the practioner.
I. The fungal cell
The fungal cell wall is a structure that is both essential for the fungus and absent from mammalian cells, and consequently may be an ideal target for antifungal agents. Inhibitors of the biosynthesis of two important cell wall components, glucan and chitin, already exist. Polyoxins and the structurally related nikkomycins (both consist of a pyrimidine nucleoside linked to a peptide moiety) inhibit chitin synthase competitively, presumably acting as analogs of the substrate uridine diphosphate (UDP)-N-acetylglucosamine (chitin is an N-acetylglucosamine homopolymer), causing inhibition of septation and osmotic lysis. Unfortunately, the target of polyoxins and nikkomycins is in the inner leaflet of the plasma membrane; they are taken up by a dipeptide permease, and thus peptides in body fluids antagonize their transport.
In most fungi, glucans are the major components that strengthen the cell wall. The glucosyl units within these glucans are arranged as long coiling chains of .beta.-(1,3)-linked residues, with occasional sidechains that involve .beta.-(1,6) linkage. Three .beta.-(1,3) chains running in parallel can associate to form a triple helix, and the aggregation of helicies produces a network of water-insoluble fibrils. Even in the chitin-rich filamentous aspergilli, .beta.-(1,3)-glucan is required to maintain the integrity and form of the cell wall (Kurtz el al. (1994) Antimicrob Agents Chemother 38:1408-1489), and, in P. carinii, it is important during the life cycle as a constituent of the cyst (ascus) wall (Nollstadt et al. (1994) Antimicrob Agents Chemother 38:2258-2265.
In a wide variety of fungi, .beta.-(1,3)-glucan is produced by a synthase composed of at least two subunits (Tkacz, J. S. (1992) in Emerging Targets in Antibacterial and Antifungal Chemotherapy (Sutcliffe and Georgopapadakou, eds), pp495-523, Chapman & Hall; and Kang et al. (1986) PNAS 83:5808-5812). One subunit is localized to the plasma membrane and is thought to be the catalytic subunit, while the second subunit binds GTP and associates with and activates the catalytic subunit (Mol et al. (1994) J Biol Chem 269:31267-31274).
Two groups of anticandidal antibiotics known in the art interfere with the formation of .beta.-(1,3)-glucan: the papulacandins and the echinocandins (Hector el al. (1993) Clin Microbiol Rev 6:1-21). However, many of the papulacandins are not active against a variety of Candida species, or other pathogenic fungi including aspergillus. The echinocandins, in addition to suffering from narrow activity spectrum, are not in wide use because of lack of bioavilability and toxicity.
II. Protein Prenylation
Covalent modification by isoprenoid lipids (prenylation) contributes to membrane interactions and biological activities of a rapidly expnanding group of proteins (see, for example, Maltese (1990) FASEB J 4:3319; and Glomset et al. (1990) Trends Biochem Sci 15:139). Either farnesyl (15-carbon) or geranylgeranyl (20-carbon) isoprenoids can be attached to specific proteins, with geranylgeranyl being the predominant isoprenoid found on proteins (Fransworth et al. (1990) Science 247:320).
Three enzymes have been described that catalyze protein prenylation: farnesyl-protein transferase (FPTase), geranylgeranyl-protein transferase type I (GGPTase-I), and geranylgeranyl-protein transferase type-II (GGPTase-II, also called Rab GGPTase). These enzymes are found in both yeast and mammalian cells (Schafer et al. (1992) Annu. Rev. Genet. 30:209-237). FPTase and GGPTase-I are .alpha./.beta. heterodimeric enzymes that share a common .alpha. subunit; the .beta. subunits are distinct but share approximately 30% amino acid similarity (Brown et al. (1993). Nature 366:14-15; Zhang et al. (1994). J. Biol. Chem. 269:3175-3180). GGPTase II has different .alpha. and .beta. subunits and complexes with a third component (REP, Rab Escort Protein) that presents the protein substrate to the .alpha./.beta. catalytic subunits. Each of these enzymes selectively uses farnesyl diphosphate or geranylgeranyl diphosphate as the isoprenoid donor and selectively recognizes the protein substrate. FPTase farnesylates CaaX-containing proteins that end with Ser, Met, Cys, Gln or Ala. GGPTase-I geranylgeranylates CaaX-containing proteins that end with Leu or Phe. For FPTase and GGPTase-I, CaaX tetrapeptides comprise the minimum region required for interaction of the protein substrate with the enzyme. GGPTase-II modifies XXCC and XCXC proteins; the interaction between GGPTase-II and its protein substrates is more complex, requiting protein sequences in addition to the C-terminal amino acids for recognition. The enzymological characterization of these three enzymes has demonstrated that it is possible to selectively inhibit one with little inhibitory effect on the others (Moores et al. (1991) J. Biol. Chem. 266:17438).
GGPTase I transfers the prenyl group from geranylgeranyl diphosphate to the sulphur atom in the Cys residue within the CAAX sequence. S Cerevisiae proteins such as the Ras superfamily proteins Rho1, Rho2, Rsr1/Bud1 and Cdc42 appear to be GGPTase substrates (Madaule el al. (1987) PNAS 84:779-783; Bender et al. (1989) PNAS 86:9976-9980; and Johnson el al. (1990) J Cell Biol 111:143-152).
III. Protein Kinase C
Members of the family of phospholipid-dependent, serine/threonine-specific protein kinases known collectively as protein kinase C (PKC) respond to extracellular signals that act through receptor-mediated hydrolysis of phosphatidylinositol-4,5-bisphosphate to diacylglycerol (DAG) and inositol-1,4,5-trisphosphate (IP.sub.3) (Hokin (1985) Annu. Rev. Biochem. 54, 205-235.). DAG serves as a second messenger to activate PKC (Takai et al. (1979) Biochem. Biophys. Res. Commun. 91, 1218-1224; Kishimoto et al. (1980) J. Biol. Chem. 255, 2273-2276; Nishizuka 1986) Science 233, 305-312; and Nishizuka (1988) Nature 334, 661-665), and IP.sub.3 functions to mobilize Ca.sup.2+ from intracellular stores (Berridge et al. (1984) Nature 312, 215-321). Twelve distinct subtypes of mammalian PKC have been reported to date (Nishizuka, Y. (1992) Science 258, 607-614; Decker et al. (1994) TIBS 19:73-77). The four initially identified isozymes, .alpha., .beta.I, .beta.II, and .gamma., are structurally closely related to each other and display similar catalytic properties.
Mammalian PKC is thought to play a pivotal role in the regulation of a host of cellular functions through its activation by growth factors and other agonists. These functions include cell growth and proliferation, release of various hormones, and control of ion conductance channels. Indirect evidence suggests that PKC induces the transcription of a wide array of genes, including the proto-oncogenes c-myc, c-fos, and c-sis, human collagenase, metallothionein II.sub.A, and the SV40 early genes.
The PKC1 gene of budding yeast encodes a homolog of the .alpha., .beta., and .gamma. isoforms of mammalian Protein Kinase C that regulates a MAPK-activation pathway. Loss of PKC1 function results in a cell lysis defect that is due to a deficiency in cell wall construction.